jhuneurofandomcom-20200214-history
Cognitive Neuroscience Lecture 4: Vision: The Computational Challenges
L4: Vision: The Computational Challenges - Vision is better than any computer’s - 55% of cortex for vision o Only 11% for touch, 3% audition The problem of vision - To recover the structure of the world from part of the electromagnetic spectrum based on shifting retinal images - Multiple possibilities of structure from the same pattern of light (3D to 2D) o Retinal projection could be from different objects ( \ | / ) Myths about vision - Vision provides a faithful record o Fraser spiral/optical illusion o Motion illusion o Size illusion - Vision is passive o Impossible figures: visual representation is coarse o Change blindness: o Troxler’s fading o Color perception (b&w to color when eyes close) - Vision is accomplished by our eyes - A lot of our brain is devoted to vision Truths about vision - Vision is tricky to study o Instinct blindness: you need to make the natural seem strange in order to ask why ''of an instinctive human act o '''Self-referencing': § up, down, left, right § how would you describe color to a blind person? § What’s it like to be a bat - Vision itself is impossible o Inverse problem: retina is 2D and world is 3D § Inverse problem: orientation and shape (circle could be a cylinder of many lengths) § Inverse problem: distance and size (the monsters chasing are an example of \ | / ) § Inverse problem: reflectance and illumination (perceive things in shadow being extra bright) o Solution: make assumptions and use them to make inference o Result: vision is an inferential, constructive process § You don’t see what’s there; you see what’s most likely to be there given your assumptions - Vision makes assumptions o Unconscious inference: infer most likely § we perceive objects most likely to produce received sensory stim § E.g. straight instead of wiggly b/c straight more likely · Avoid “accidental viewpoint” o Coincidence avoidance § E.g. kanizsa figures, idesawa’s spike sphere · Interpreted as sphere with spikes b/c more likely than independent spikes o Pragnanz (succinctness, simplicity) § Assumes simplest thing is correct · E.g. square with plus, square with spike o Lighting is uniform, shadows are smooth, and local info matters most § E.g. cylinder with one checker in shadow, one in light: both are same color o Shadows move with object § E.g. Sphere moving above checkerboard o Unchanging is uninformative § That’s why you don’t perceive blood vessels on your retina o Faces are convex § Spinning mask o Vision fills in the blanks § Blind spot: where optic nerve feeds back into brain Case study: depth perception How do we see the third dimension? Inverse problem Depth cues: '''cues that support depth perception in 3-D moving world - Monocular cues - Binocular cues - Dynamic cues - Pictorial cues '''Accomodation: degree of strain on the lens—'not much use beyond 1m' MONOCULAR - Far: thin lens - Near: thick lens Convergence: relative angle between eyes—'not much use beyond 1m '''MONOCULAR - Far: small angle of convergence - Near: large angle of convergence '''Motion parallax': relative motion of objects MONOCULAR - Distant objects move least, close objects move most Binocular disparity: having two eyes - Motion parallax requires moving head - Accommodation and convergence require moving eyes - Binocular disparity does not require moving head - The amount of disparity differs depending on distance between two objects Pictorial Cues: '''cues that support depth perception in flat, static images - '''Occlusion: one object is covering the other o T junctions '''imply occlusion, and thus a depth ranking (BUT NOT MEASUREMENT) - '''Shadows: depth ranking, relationship between objects and surfaces - Linear perspective: 'parallel lines must converge in the distance - '''Relative height (horizon) ': Height in field relative to horizon o We perceive objects near the horizon as more distant - '''Known size: if a known object has a ‘canonical size’, we can determine distance based on retinal input o Often overruled by other cues o E.g. child looks big b/c standing close to you o Top-down knowledge - Texture gradient: assuming ground is uniform, changes in spatial scale reflect distance - Atmospheric perspective: more intervening atmosphere at distance increases, growing haze o Blue mountains with farther back ones light blue, hazy Constraint satisfaction: putting it all together - Cues arise from depth distances in 3D world - Each cue in a local region gives rise to a set of possible interpretations - Each possibility constrains assignment of edges and surfaces around cue - Final interpretation is what is most compatible with '''all '''cues (like Sudoku!) Conflict of depth cues - E.g. ames room: assume corners meet at 90 deg. But not necessarily true